The foregoing related applications describe certain background aspects and problems associated with current systems and methods that may pertain to the current invention. The reader is referenced to this information in addition to the following background information.
Hereinafter, the term “human entity” generally refers to one or more human beings or organizations comprising humans. The term “human shopping” generally refers to the process of seeking out, matching up and/or selection of human entities. The term “manage” is also used herein and generally refers to the sequencing, coordinating, tracking and reporting on of various aspects of the system and method of the current invention.
3.1 Management Functions
Services currently exist that use automation to help to sequence, coordinate, track and report on the status of the activities involved in the seeking out, matching up and/or selecting of human candidates. Examples of such current services pertain to the process of hiring prospective employees and dating services involving the process of matching people seeking personal relationships. For example, Monster.Com, an Internet-based employment service and job posting board, offers a service called “Momentum.” This service was purchased from another company and has been available in some form for several years. Entities which may seek out humans on the Momentum service typically include employers looking for employees, as well as companies such as employee search firms.
Among its various functions, the Momentum service keeps track of the status of the hiring process for job positions offered by its employer human shoppers. It separately tracks the status of each job-seeker being considered for each open employment position. In addition, it allows companies seeking out prospective employees to have more than one open position tracked by the system at the same time. The status information available to a company seeking out prospective employees may therefore consist of status information about several open positions offered by that company.
The Momentum service also offers a calendar scheduling system that facilitates the selection of times for interviews and other meetings between job-seekers and representatives of the prospective employers. In addition, it offers a central repository for interviewers and other staff members to post notes and comments regarding a job-seeker. The Momentum service also offers a central repository for storing e-mail and other messages regarding the hiring process, making it easier for prospective employer's staff members to review such messages.
To summarize, the Momentum service uses the Internet and automation to track the status of the hiring process, report on the status of the hiring process, and store and publish information about the hiring process at a central location. Other services and systems currently available offer a subset of these features, with variations of the same basic ideas.
Despite the level of automation used, the Momentum service, and other existing services and programs referred to in the previous paragraph, require human intervention to manage the process of completing the various steps in the hiring process. As mentioned above, the “management” requiring human intervention may involve sequencing, coordinating, tracking and reporting on the status of the various steps in the hiring process.
For example, in such existing services and programs, if an interview is possible at a given point in the hiring process, a human must typically: 1) decide that an interview is needed, 2) determine who are the individuals that must be present at the interview, 3) decide when and where the interview is to take place, 4) send notes or messages to the various individuals in order to decide when and where the interview is to take place and to notify the participants of the selected time and place, 5) determine which set of questions are to be presented to the job-seeker during the interview, and 6) make sure that the results of the interview are stored or posted in the proper location so that a decision maker can find them when needed.
Even for steps in the hiring process that do not involve any human beings to carry out the major parts of the step, a human is often needed in existing services to carry out some parts of the step. For example, in an employment service, it may be necessary to carry out background checks including credit checks, criminal checks and D&B checks. To use one of these as an example, if a credit check on a job-seeker is needed at a certain point in the hiring process, although an automated system may be used for communication, storage and display of information, a human typically must carry out management functions such as 1) requesting the credit report from one or more credit bureaus, 2) requesting the credit report again if it does not arrive after a pre-determined amount of time and 3) placing the relevant contents of the credit report in the proper location in the automated system so that a decision maker can find it when needed.
At a higher level, a human is typically required in all existing services to 1) determine the correct or desired sequence in which various steps are to be executed, 2) make sure that the various steps executed are in fact executed in the desired sequence, and 3) make sure that no essential steps are left out. Further a human is typically required in all existing services to 1) make sure that the completion (or failure to complete) a step is properly recorded for others to view, 2) actually record the results of each step for others to view, 3) communicate status information about the hiring process for each job-seeker and for each order to the proper persons at the proper times and 4) notify the proper individuals if something has gone wrong, thus requiring higher-level human intervention.
In summary, the foregoing illustrates that at least one human being is needed in order to carry out many functions of the existing systems and methods for human shopping such as seeking out, matching up and/or selecting. This represents a significant drawback of existing systems and methods because the many tasks to be tracked by humans are susceptible to human error. Such errors could lead to inefficiencies in both how the system operates as well as the cost involved in operating the system.
As noted in the previous paragraph, even though services and programs that use automation to facilitate the hiring process exist, they all rely upon human intervention to manage the steps in the process. This is also true of matchmaking and dating services, as well as services that help in shopping for expert consultants, where multiple steps are involved in the shopping and selection process. In other words, all known systems that involve sequencing, coordinating, tracking and reporting on status of human shopping and selection processes use human beings in these management functions. None of these systems effectively automate, with little or no human intervention, the management functions in human shopping and selecting processes. Reasons for this may include the following:
First, the people that created these systems typically believe that human judgement is needed in order to make many of the lower-level decisions related to the shopping and selection of human beings or human organizations, such as interpreting the results of interviews or testing. Accordingly, these same people generally conclude that humans are also needed for the management functions. Second, humans have always played the management role, therefore it is concluded that humans must still be needed in this role (i.e social inertia). Third, there have been insufficient economic incentives to eliminate humans from the management role. Fourth, it is not clear that automation can successfully carry out the management role.
Whatever the reason or combination of reasons, automation has not been used in the management role except in the limited role to assist a manager in handling communications and record keeping. No one has attempted to substantially or completely automate the management of the process of selecting humans.
3.2 Scoring and Ranking
When considering the steps that may occur in the overall human matching, selection or shopping process, certain steps may involve the ranking or scoring of human candidates with respect to selection criteria. Current systems involving employment and dating services include these steps.
Typically, there are three parties to such services: two of these parties each provide a description of the kind of person, organization, position or job they are each looking for and/or a description of relevant aspects of themselves. These parties may generally be referenced herein as “shoppers.” A third party is a service that attempts to score and/or rank the degree to which pairs of shoppers match each other.
Distinctions exist between shopping or matching services that seek human entities (and the roles or functions which human entities perform) and shopping services that seek non-human objects such as books or other products. It should also be noted that a shopping or matching service may also seek services provided by humans, but may still treat the services as if they are non-human entities, i.e. by not taking into consideration personal and social characteristics regarding the services.
Services that deal with non-human entities may generally not require the same level of complexity or subtlety in the methods used to locate, rank or score entities to be selected. This is partly because (1) non-human entities are subject to more tangible and quantifiable description, (2) measurement of human personal and social characteristics such as intelligence, skills, attitudes, values and character is fundamentally imprecise and changes over time, (3) requirements in the areas of human personal and social characteristics is more dynamic for each shopper, i.e. each shopper may frequently alter their requirements over time and (4) the consequences of an error in selecting such non-human entities may be generally less than selecting human entities.
A distinction can be made between a human shopping service that deals with “static” vs. “dynamic” characteristics. In an employment service, for example, the process of locating and selecting a secretary is considerably less complex than the process of locating and selecting a computer programmer expert in the field of wireless devices. This is partially because the requirements for a secretary are relatively static, whereas the requirements for the computer programmer may change every month as the technology evolves, and may thus be dynamic.
A human shopping service or process may be generally described as “symmetrical” or “non-symmetrical.” A symmetrical service or process is one in which human entities on both sides of the match sought to be made, use the same or substantially the same characteristics and methods to locate, score, rank and/or select appropriate human entities on the other side. Examples of symmetrical services are dating, matchmaking and barter services.
A non-symmetrical service is one in which human entities on each side of the match to be made use different (or a mix of different and the same) characteristics and methods to locate, score, rank and/or select human entities on the other side. Typically, though not necessarily, in a non-symmetrical service or process, human entities on one side offer to pay for services provided by the human entities on the other side. Examples of non-symmetrical services or processes are employment services (matching employers with employees), contractor services (matching contractors such as building contractors with a sub-contractor), professional locator services (in which patients find doctors or human shoppers find lawyers), talent agencies (in which studios find actors or event planners find entertainers), and so on.
In non-symmetrical services, the party which is offering to pay to engage the other party may be more interested in the personal and social characteristics of the party offering to perform a service, than the other way around. For example, in an employment service, companies seeking employees are likely to be more interested in the intelligence and attitudes of people seeking employment, than the other way around. However, this is not necessarily the case. Even in a symmetrical service, the nature of the information acquired and stored about each side need not be exactly the same.
Hereinafter, the term “selection criteria set” generally refers to a description of whatever it is a human shopper is looking for and/or a description of the relevant characteristics of that human shopper him or herself, i.e., the characteristics of that human shopper which may be evaluated by human entities with which that human shopper may engage. Some human shopping services make an explicit distinction between these two kinds of selection criteria.
For example, a dating service may accept and store separately: (1) a description of what kind of person each human shopper is looking for and (2) a description of each human shopper's relevant personal attributes. For example, a service may accept and store separately the fact that a shopper likes moonlit walks on the beach as well as the fact that he/she is looking for others that like moonlit walks on the beach, i.e., separate information about the shopper him/herself and the person the shopper seeks. A different service may not make this distinction clearly and may store simply “interested in moonlit walks on the beach” without clearly distinguishing whether this information pertains to the shopper or the person sought by the shopper.
In many human shopping services, a human shopper may be permitted to submit more than one “selection criteria set”. That is, a human shopper may provide different descriptions of him or herself or different descriptions of the entity sought, e.g. one or more types of jobs or dates, that the human shopper can offer or is looking for, permitting a single human shopper to be looking for more than one type of entity.
For example, a person looking for a job may be qualified for more than one type of job. Similarly, companies may be looking for more than one type of employee such as where the employer needs to fill more than one type of job position at any given time. Each job position may have a different selection criteria set, i.e. a different set of requirements and skills, including personal characteristics and character traits, as to what kind of employees would best fill that job and/or a different description of the job itself.
In the employment service example, two jobs offered may be identical in every way except that in one position an employer seeks a “dependent” type of person whereas in another position that employer seeks an “independent” type of person. Similarly, an individual human shopper seeking employment may provide more than one selection criteria set, i.e. more than one description of him/herself (emphasizing different skill sets), along with descriptions of more than one type of job that she or he can perform. Many services explicitly permit multiple selection criteria sets for each human shopper, but request and store general information about each human shopper only one time, so that the human shopper does not have to enter general information about itself more than one time.
Human shopping services use various different formats for describing selection criteria sets, including for example: (1) free text, (2) multiple-choice and/or (3) audio/visual. An example of free text is a resume submitted by an individual looking for employment. An example of multiple-choice is a questionnaire or checklist in which acceptable answers are limited to those provided in or with the questionnaire or checklist. An example of audio/visual is a photograph of an individual human shopper or a working environment, or a recording of an interview. Many human shopping services allow selection criteria sets to consist of information in more than one of these forms.
One problem with existing human shopping systems is as follows. When multiple-choice is used in a selection criteria set, there is typically just one or a very small number of different sets of questions provided by the human shopping service. For example, in a dating service, a different set of questions may be provided for a human shopper seeking a casual relationship as opposed to a human shopper seeking marriage or serious relationship. As an another instance, in an employment service, a different set of questions may be provided when an employer seeks to fill a technical job than when an employer seeks to fill a management or clerical job. However, in both cases, further detail is not requested or considered.
For example, in the case of an employment service within the category of technical jobs, the kinds of questions and multiple-choice options relevant to a systems programming job may be substantially different than the kinds of questions relevant to a systems administration job. However, the questionnaire actually provided by a certain service may be the same for both positions.
As a further example, in a matchmaking service, a human shopper seeking a serious relationship may be interested in marriage or just a non-romantic friendship, yet the service may not request or store such preferences. Similarly, in a matchmaking service, among individuals that state an interest in marriage, the kinds of questions relevant to a religious individual may be substantially different than those relevant to a non-religious individual. Yet both categories of human shoppers (religious and non-religious) may be presented with the same set of questions since they have the same stated purpose (marriage).
Further, there may be only a small number of multiple-choice options within each multiple-choice question. For example, in an employment service, in hiring a computer programmer, the service may ask in a multiple-choice question, “Please indicate the programming methods you are experienced with from among the following list”, but may not include methods that have been developed in the last 6 months, such as Aspect programming or Extreme Programming.
Thus, one of the problems with existing services is that, even if there are several different sets of questions asked for different purposes, the number of different sets of questions is not large enough to reach highly useful conclusions. In sum, the use of a small number of different sets of questions and small number of different options within questions generates conclusions that are of limited usefulness to the human shoppers. It can be appreciated how this will lead to inefficiencies and/or errors in the matching of human shoppers and the humans they are seeking.
Before continuing, it will be helpful to comment on what it is meant by degrees of “useful”, in the context of a human shopping service. In a human shopping service, in order for the conclusion reached or suggested by the service to be highly useful, the service is preferably capable of selecting human entities which turn out to be good selections most or some acceptable percentage of the time. Inappropriate recommendations may waste the time and resources of human shoppers. To put this point more precisely, the frequency at which a human shopping service is believed by human shoppers to generate inappropriate selections or recommendations may determine the amount of additional interviewing and/or testing that the human shoppers believe they have to perform outside of what is provided by the service.
As discussed above, a human shopping service that uses only one or a small number of question sets and a small number of multiple-choice options may not generate highly useful results. This is because the particular sets of questions and multiple-choice options available do not exactly or at least efficiently serve each human shopper's purpose.
There are some employment services which do maintain a large number of different question sets and multiple-choice options. For example, one service asks a job-seeker who is interested in a technical job to select (from a multiple-choice list) what job category the job-seeker is interested in, such as “network administration”, “web site development”, etc. Once the job-seeker makes this selection, the service asks the job-seeker to select (from a multiple-choice list) one or more particular technology skills that the job-seeker possesses. This multiple-choice list is different depending on the job category selected earlier.
This requires that the service create a database of job categories and technology skills, with each job category related to one or more technology skill and with each technology skill related to one or more job categories. A database (generally referred to herein as the “knowledge-base”) of this kind may have more than two levels. For example, a four-level knowledge-base might consist of job categories (e.g. network administration), technology skills (e.g. routers), specific brand skills (e.g. Cisco-brand routers), and specific model skills (e.g. Cisco 5300 routers). It should be noted that reference to particular manufacturers and products herein is by example only. There is no theoretical limit to the degree of specificity and number of levels in this knowledge-base.
Because technology is continuously changing, technology skills and job categories and the relationships between them are continuously evolving. This requires that the service update the knowledge-base of technology skills and job categories continuously, a costly and difficult effort since it requires detailed and current knowledge of all the areas of job categories and related technology skills. Furthermore, the process of updating may require substantial human input to actually provided the updated information into the system. And this human input would typically require the involvement of human experts that have the updated information in the first place.
In an automated system, the process of updating may entail the use of “expert system” technology. An “expert system” as that term is used herein, generally refers to an automated system which attempts to emulate the functioning of a human expert in a given field to solve a problem, diagnose the cause of a condition, schedule an event or recommend action. In a typical expert system, the method by which the conclusion is reached is difficult to define rigorously. Examples in current use are automated systems that perform medical diagnoses, recommend repair procedures for complex machines such as airplanes and power plants or make credit-granting decisions.
Expert systems generally fall into two very broad categories: (1) “forward chaining” or forecasting, and (2) “backward chaining” or diagnosing. In a “forward chaining” expert system, the expert system is used to predict an outcome or event. An example is a logistics system, used to predict the best possible routes for vehicles moving large numbers of shipments between large numbers of locations. In a “backward chaining” expert system, the expert system is used to determine the cause of an event or diagnose a condition. An example is a medical diagnosis system, used to determine the cause of symptoms displayed by a patient.
There are various characteristics common to most expert systems.
1) Expert systems are often characterized by a process wherein data is input in many stages, and wherein the particular data input that the system requests or gathers at any particular stage is dependent upon data input to the system in prior stages. For example, an expert system performing medical diagnoses might first request known symptoms exhibited by the patient. Based upon the particular combination of symptoms received by the system, the system may then request that certain tests be performed upon the patient. The results of these tests are then input to the system. The system may then combine the results of these tests with the original group of symptoms and arrive at a diagnosis, or may then request additional tests or information before a diagnosis is determined.
2) Expert systems typically make use of a stored database of information about the field of knowledge in general and possibly about the history of the particular individual (person, airplane, etc.) under examination. This information can be used both in the process of determining what additional information is needed at interim stages of the decision-making process, and in the process or reaching a final conclusion.
3) Expert systems often reach conclusions that are not certain. In light of this fact, an expert system may output a probability that a conclusion is correct and/or may output multiple possible conclusions.
4) Expert systems are often designed to learn from experience. In other words, after the system outputs a conclusion, the results of applying that conclusion are later input back to the system and stored as part of the system's database. These results may then be used by the system to modify future conclusions in similar cases.
3.3 Interviewing and Testing
Current systems and methods that seek to match human entities, e.g., a dating service that matches two individuals for a date or an employment service that matches prospective employers and employees, generally rely on the information provided by each human entity in order to perform the match. As can be appreciated, the success of the match generally relies on the accuracy of the information provided by the human entities.
In such systems or methods, problems arise where the information provided is not accurate. Inaccurate information may be provided inadvertently or intentionally. For example, a form or other means used for a human shopper to input information into a system may be mistakenly filled out. Also, an individual may misrepresent his or her characteristics in hopes of being matched with a “better” date or a prospective employee may misrepresent his or her qualifications in hopes appearing more qualified to get a better job. Either way, the resulting match is not desirable or at least is not as beneficial as it otherwise should be.
Accordingly, there is a need to verify the accuracy of information provided by human shoppers. There is also a need to correct the information where necessary. Process that may be used to address such needs are generally referred to as “inspection and verification” processes. Current systems and methods for human shopping may utilize testing and interviewing in order to carry out inspection and verification. Such testing or interviewing may indeed expose inaccurate information. For example, an interview may bring to light inaccuracies contained in a job application or resume. As another example, videotapes of individuals seeking a date are used by dating services to provide better information than that which might be contained in a written questionnaire profiling the person.
The process of testing and interviewing typically used, however, is normally an inexact process for inspection and verification, since the processes typically used do not attempt to accurately inspect and verify every claim made or piece of information given by a human shopper. For example, the same general purpose test may be given to many employment job-seekers who differ significantly in the claims they have made and the jobs for which they are applying. In this case, it can be appreciated that one single test will likely not reveal inaccuracies that may exist in information provided by various people. But to customize the test for each employee and each job position would generally require too great an effort.
Similarly, in the same context, when an interview is carried out during the hiring process, the interviewer may be able to concentrate on the particular claims of the job-seeker and the job requirements. However, there is no assurance that any particular interviewer is expert enough to properly evaluate the claims of the job-seeker on every point that may be relevant and/or necessary.
Accordingly there is a need for automation in the testing and/or interviewing of human entities so that interviewing or other verification procedures may be more efficiently used. There is also a need for an interview or testing procedure that is “intelligent” or otherwise based on a knowledge-base of questions and answer options that reflect expertise in a given area. In this way, the information or claims made by a human entity may be more effectively verified and/or corrected.
And because technology and human culture are continuously changing, there is a need for any such knowledge-base of questions and answer options to continuously evolve for the system to be useful. For example, in an employment service, this may require that the service update the knowledge-base of technology questions and answer options continuously so that prospective employees are asked questions that reflect the current state of technology that may be desired by the employer. To this end, there is a need for the efficient and cost-effective updating of any such knowledge-base.